On the role of feedback in visual processing: a predictive coding perspective
Not on the map. Why?
Abstract (arXiv)
Brain-inspired machine learning is gaining increasing consideration, particularly in computer vision. Several studies investigated the inclusion of top-down feedback connections in convolutional networks; however, it remains unclear how and when these connections are functionally helpful. Here we address this question in the context of object recognition under noisy conditions. We consider deep convolutional networks (CNNs) as models of feed-forward visual processing and implement Predictive Coding (PC) dynamics through feedback connections (predictive feedback) trained for reconstruction or classification of clean images. To directly assess the computational role of predictive feedback in various experimental situations, we optimize and interpret the hyper-parameters controlling the network's recurrent dynamics. That is, we let the optimization process determine whether top-down connections and predictive coding dynamics are functionally beneficial. Across different model depths and architectures (3-layer CNN, ResNet18, and EfficientNetB0) and against various types of noise (CIFAR100-C), we find that the network increasingly relies on top-down predictions as the noise level increases; in deeper networks, this effect is most prominent at lower layers. In addition, the accuracy of the network implementing PC dynamics significantly increases over time-steps, compared to its equivalent forward network. All in all, our results provide novel insights relevant to Neuroscience by confirming the computational role of feedback connections in sensory systems, and to Machine Learning by revealing how these can improve the robustness of current vision models.
Key facts. Cited by 2 papers in the map (data through 7 Oct 2026). Builds on 22 of its 48 references that are in the map. First posted on arXiv 8 Jun 2021.
Builds on
22 of 48 references.
References · 22
- 2014Adam: A Method for Stochastic Optimizationcited by 36,960
- 2014Explaining and Harnessing Adversarial Examplescited by 7,287
- 2019EfficientNet: Rethinking Model Scaling for Convolutional Neural Networkscited by 6,415
- 2017Towards Deep Learning Models Resistant to Adversarial Attackscited by 6,326
- 2013Intriguing properties of neural networkscited by 5,114
- 2019Benchmarking Neural Network Robustness to Common Corruptions and Perturbationscited by 2,857
- 2014Deep Neural Networks are Easily Fooled: High Confidence Predictions for Unrecognizable Imagescited by 774
- 2018Defense-GAN: Protecting Classifiers Against Adversarial Attacks Using Generative Modelscited by 362
- 2017MagNet: a Two-Pronged Defense against Adversarial Examplescited by 268
- 2018Learning long-range spatial dependencies with horizontal gated-recurrent unitscited by 98
- 2019Recurrence is required to capture the representational dynamics of the human visual systemcited by 58
- 2017APE-GAN: Adversarial Perturbation Elimination with GANcited by 57
10 more
- 2017The Robust Manifold Defense: Adversarial Training using Generative Modelscited by 35
- 2016Feedback Networkscited by 29
- 2018Task-Driven Convolutional Recurrent Models of the Visual Systemcited by 27
- 2018Deep Predictive Coding Network for Object Recognitioncited by 23
- 2020Neural Networks with Recurrent Generative Feedbackcited by 23
- 2021Predify: Augmenting deep neural networks with brain-inspired predictive coding dynamicscited by 19
- 2013Deep Predictive Coding Networkscited by 13
- 2021Predictive coding feedback results in perceived illusory contours in a recurrent neural networkcited by 10
- 2019Recurrent Feedback Improves Feedforward Representations in Deep Neural Networkscited by 5
- 2021GAttANet: Global attention agreement for convolutional neural networkscited by 2
Built on by
Citing papers per year: 2024: 1; 2025: 1.
2 papers.